Wells Fargo, JPM, Citi, Bank of America share this year’s AI productivity gains

Defense and Compliance Attorneys

Executives from Wells Fargo, JPMorgan Chase, Citigroup and Bank of America are reporting measurable AI-driven productivity gains in areas like software development, operations, and customer service, often in the double-digit percentage range. These gains are allowing the banks to “do more with the same or fewer people” and are expected to shape headcount and cost structures over the next few years.

Key 2025 AI productivity metrics

JPMorgan Chase: Productivity in consumer and community banking has risen from about 3% to roughly 6% annual efficiency gains after one year of scaled AI deployment, with some operational roles expected to see 40% to 50% productivity improvements in coming years. The bank anticipates being able to process more business volume with fewer incremental costs as AI automates routine operational tasks.
Wells Fargo: Generative AI tools have made software developers roughly 30% to 35% more productive, which leadership characterizes as a “real efficiency” gain that boosts output without adding staff. Management has indicated that AI offers a significant opportunity to raise productivity and will likely reduce staffing needs over time, even though current headcount has been held roughly flat.
Citigroup: Citi reports about a 9% uplift in coding productivity from AI-assisted development tools, improving both internal software delivery and customer self-service capabilities. The bank is embedding AI copilots into employee workflows to further extend these gains beyond technology teams.
Bank of America: BofA’s virtual assistant Erica, which handles a very large volume of customer interactions, is described internally as equivalent to roughly tens of thousands of full-time employees in terms of handled service load, materially reducing branch and call-center pressure. The bank has also disclosed around 20% efficiency gains among developers using generative AI tools, alongside multi-year plans to invest billions of dollars into AI to boost banker productivity and revenue per employee.

Summary table of reported gains

Bank	Main AI use cases (2025)	Reported productivity impact
JPMorgan Chase	Ops automation, consumer banking, service workflows	Annual efficiency gains doubled from ~3% to ~6%; select ops roles targeted for 40–50% productivity improvement.
Wells Fargo	Generative AI for coding, internal ops, compliance	Developer productivity up about 30–35%; broader “do more with same staff” gains across operations.
Citigroup	AI copilots, coding, customer self-service	Around 9% coding productivity uplift; improving self-service and assisted call productivity.
Bank of America	Erica virtual assistant, gen-AI for developers	Erica seen as equivalent to roughly 11,000 FTEs in workload terms; about 20% developer efficiency gain and large planned AI investment.

Implications for jobs and costs

Leaders across these banks emphasize that AI is primarily a productivity and cost-efficiency lever, but they also acknowledge it will ultimately mean fewer roles in some functions as automation scales. At the same time, they are increasing hiring for AI and data specialists, suggesting a shift in workforce composition rather than a uniform reduction in total employment.